| { | |
| "best_metric": 27.650982108014144, | |
| "best_model_checkpoint": "./checkpoint-7000", | |
| "epoch": 13.063071428571428, | |
| "eval_steps": 500, | |
| "global_step": 7000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 4.8e-08, | |
| "loss": 1.3628, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.8e-08, | |
| "loss": 1.3146, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.4799999999999998e-07, | |
| "loss": 1.3827, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.98e-07, | |
| "loss": 1.288, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.48e-07, | |
| "loss": 1.2545, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.98e-07, | |
| "loss": 1.2401, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.4799999999999994e-07, | |
| "loss": 1.2529, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.98e-07, | |
| "loss": 1.25, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.48e-07, | |
| "loss": 1.1481, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.979999999999999e-07, | |
| "loss": 1.0013, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 5.48e-07, | |
| "loss": 0.9133, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 5.979999999999999e-07, | |
| "loss": 0.9555, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 6.48e-07, | |
| "loss": 0.8325, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 6.979999999999999e-07, | |
| "loss": 0.7611, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 7.48e-07, | |
| "loss": 0.7607, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 7.98e-07, | |
| "loss": 0.6794, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 8.48e-07, | |
| "loss": 0.6333, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 8.98e-07, | |
| "loss": 0.6445, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.479999999999999e-07, | |
| "loss": 0.6742, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.98e-07, | |
| "loss": 0.739, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 0.7505670785903931, | |
| "eval_runtime": 1365.9553, | |
| "eval_samples_per_second": 11.817, | |
| "eval_steps_per_second": 0.37, | |
| "eval_wer": 35.00880999964524, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.963076923076924e-07, | |
| "loss": 0.6733, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.924615384615386e-07, | |
| "loss": 0.6337, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.886153846153845e-07, | |
| "loss": 0.6459, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.847692307692307e-07, | |
| "loss": 0.6253, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.809230769230769e-07, | |
| "loss": 0.6447, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.77076923076923e-07, | |
| "loss": 0.6424, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.732307692307692e-07, | |
| "loss": 0.6403, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.693846153846154e-07, | |
| "loss": 0.6501, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.655384615384616e-07, | |
| "loss": 0.619, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.616923076923077e-07, | |
| "loss": 0.6014, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.578461538461537e-07, | |
| "loss": 0.581, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.539999999999999e-07, | |
| "loss": 0.561, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.501538461538461e-07, | |
| "loss": 0.5794, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.463076923076922e-07, | |
| "loss": 0.6063, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.424615384615384e-07, | |
| "loss": 0.5932, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.386153846153846e-07, | |
| "loss": 0.5755, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.347692307692308e-07, | |
| "loss": 0.5826, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.309230769230769e-07, | |
| "loss": 0.6204, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 9.270769230769231e-07, | |
| "loss": 0.5591, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 9.232307692307692e-07, | |
| "loss": 0.6131, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 0.6595470905303955, | |
| "eval_runtime": 1267.4973, | |
| "eval_samples_per_second": 12.735, | |
| "eval_steps_per_second": 0.398, | |
| "eval_wer": 31.429821553161545, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.193846153846154e-07, | |
| "loss": 0.5829, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.155384615384615e-07, | |
| "loss": 0.5177, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.116923076923077e-07, | |
| "loss": 0.5542, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.078461538461538e-07, | |
| "loss": 0.591, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.039999999999999e-07, | |
| "loss": 0.5383, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.001538461538461e-07, | |
| "loss": 0.5493, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.963076923076923e-07, | |
| "loss": 0.5522, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 8.924615384615384e-07, | |
| "loss": 0.5528, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 8.886153846153845e-07, | |
| "loss": 0.5702, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 8.847692307692307e-07, | |
| "loss": 0.5457, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 8.809230769230769e-07, | |
| "loss": 0.5805, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 8.770769230769231e-07, | |
| "loss": 0.5246, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 8.732307692307692e-07, | |
| "loss": 0.5805, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 8.693846153846154e-07, | |
| "loss": 0.541, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 8.655384615384616e-07, | |
| "loss": 0.5333, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.616923076923076e-07, | |
| "loss": 0.5509, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.578461538461538e-07, | |
| "loss": 0.5224, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.539999999999999e-07, | |
| "loss": 0.5498, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 8.501538461538461e-07, | |
| "loss": 0.501, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 8.463076923076922e-07, | |
| "loss": 0.5311, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 0.6301009654998779, | |
| "eval_runtime": 1277.0428, | |
| "eval_samples_per_second": 12.639, | |
| "eval_steps_per_second": 0.395, | |
| "eval_wer": 30.623322256778966, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.424615384615384e-07, | |
| "loss": 0.62, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 8.386153846153846e-07, | |
| "loss": 0.5652, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 8.347692307692308e-07, | |
| "loss": 0.5905, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 8.309230769230768e-07, | |
| "loss": 0.5598, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.27076923076923e-07, | |
| "loss": 0.5169, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.232307692307692e-07, | |
| "loss": 0.5144, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.193846153846154e-07, | |
| "loss": 0.5033, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.155384615384615e-07, | |
| "loss": 0.5355, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.116923076923077e-07, | |
| "loss": 0.5146, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.078461538461539e-07, | |
| "loss": 0.5071, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.04e-07, | |
| "loss": 0.5108, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.001538461538461e-07, | |
| "loss": 0.4895, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 7.963076923076922e-07, | |
| "loss": 0.4777, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 7.924615384615384e-07, | |
| "loss": 0.5107, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 7.886153846153845e-07, | |
| "loss": 0.5348, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.847692307692307e-07, | |
| "loss": 0.4912, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.809230769230769e-07, | |
| "loss": 0.5226, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.770769230769231e-07, | |
| "loss": 0.5073, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 7.732307692307692e-07, | |
| "loss": 0.4799, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 7.693846153846153e-07, | |
| "loss": 0.551, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "eval_loss": 0.6141447424888611, | |
| "eval_runtime": 1258.2269, | |
| "eval_samples_per_second": 12.828, | |
| "eval_steps_per_second": 0.401, | |
| "eval_wer": 29.781937726901837, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 7.655384615384615e-07, | |
| "loss": 0.5604, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 7.616923076923077e-07, | |
| "loss": 0.4793, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 7.578461538461538e-07, | |
| "loss": 0.4757, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 7.54e-07, | |
| "loss": 0.501, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 7.501538461538462e-07, | |
| "loss": 0.5085, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 7.463076923076924e-07, | |
| "loss": 0.5199, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 7.424615384615385e-07, | |
| "loss": 0.5725, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.386153846153845e-07, | |
| "loss": 0.5193, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.347692307692307e-07, | |
| "loss": 0.5119, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.309230769230768e-07, | |
| "loss": 0.5094, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 7.27076923076923e-07, | |
| "loss": 0.5565, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 7.232307692307692e-07, | |
| "loss": 0.5109, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 7.193846153846154e-07, | |
| "loss": 0.4878, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 7.155384615384615e-07, | |
| "loss": 0.4965, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 7.116923076923077e-07, | |
| "loss": 0.482, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 7.078461538461538e-07, | |
| "loss": 0.4418, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 7.04e-07, | |
| "loss": 0.4697, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 7.001538461538461e-07, | |
| "loss": 0.4584, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 6.963076923076923e-07, | |
| "loss": 0.4697, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 6.924615384615385e-07, | |
| "loss": 0.4588, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "eval_loss": 0.6003413200378418, | |
| "eval_runtime": 1263.8901, | |
| "eval_samples_per_second": 12.771, | |
| "eval_steps_per_second": 0.4, | |
| "eval_wer": 29.252746473043768, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 6.886153846153846e-07, | |
| "loss": 0.5057, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 6.847692307692307e-07, | |
| "loss": 0.4379, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 6.809230769230769e-07, | |
| "loss": 0.4401, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 6.770769230769231e-07, | |
| "loss": 0.4427, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 6.732307692307691e-07, | |
| "loss": 0.4519, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 6.693846153846153e-07, | |
| "loss": 0.4947, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 6.655384615384615e-07, | |
| "loss": 0.4607, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 6.616923076923077e-07, | |
| "loss": 0.4355, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 6.578461538461538e-07, | |
| "loss": 0.4545, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 6.54e-07, | |
| "loss": 0.4304, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.501538461538462e-07, | |
| "loss": 0.5341, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.463076923076924e-07, | |
| "loss": 0.5021, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.424615384615384e-07, | |
| "loss": 0.4787, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 6.386153846153846e-07, | |
| "loss": 0.4878, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 6.347692307692308e-07, | |
| "loss": 0.474, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 6.309230769230768e-07, | |
| "loss": 0.4823, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 6.27076923076923e-07, | |
| "loss": 0.4788, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 6.232307692307692e-07, | |
| "loss": 0.457, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 6.193846153846154e-07, | |
| "loss": 0.4343, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 6.155384615384615e-07, | |
| "loss": 0.4163, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "eval_loss": 0.5935855507850647, | |
| "eval_runtime": 1271.4925, | |
| "eval_samples_per_second": 12.695, | |
| "eval_steps_per_second": 0.397, | |
| "eval_wer": 29.02924446862103, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 6.118461538461538e-07, | |
| "loss": 0.4411, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 6.079999999999999e-07, | |
| "loss": 0.5187, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 6.041538461538461e-07, | |
| "loss": 0.521, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 6.003076923076923e-07, | |
| "loss": 0.5194, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 5.964615384615385e-07, | |
| "loss": 0.5188, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 5.926153846153846e-07, | |
| "loss": 0.5057, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 5.887692307692308e-07, | |
| "loss": 0.4798, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 5.849230769230769e-07, | |
| "loss": 0.4956, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 5.810769230769231e-07, | |
| "loss": 0.4896, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 5.772307692307692e-07, | |
| "loss": 0.5033, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 5.733846153846154e-07, | |
| "loss": 0.5011, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 5.695384615384615e-07, | |
| "loss": 0.4566, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 5.656923076923076e-07, | |
| "loss": 0.4661, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 5.618461538461538e-07, | |
| "loss": 0.4607, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 5.58e-07, | |
| "loss": 0.4659, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 5.541538461538461e-07, | |
| "loss": 0.4577, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5.503076923076922e-07, | |
| "loss": 0.4721, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5.464615384615384e-07, | |
| "loss": 0.469, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5.426153846153846e-07, | |
| "loss": 0.4759, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 5.387692307692308e-07, | |
| "loss": 0.4553, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "eval_loss": 0.5837779641151428, | |
| "eval_runtime": 1248.1855, | |
| "eval_samples_per_second": 12.932, | |
| "eval_steps_per_second": 0.405, | |
| "eval_wer": 28.479949859867794, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 5.349230769230769e-07, | |
| "loss": 0.5038, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 5.310769230769231e-07, | |
| "loss": 0.5066, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 5.272307692307693e-07, | |
| "loss": 0.4496, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 5.233846153846154e-07, | |
| "loss": 0.4628, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 5.195384615384615e-07, | |
| "loss": 0.4837, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 5.156923076923077e-07, | |
| "loss": 0.4728, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 5.118461538461538e-07, | |
| "loss": 0.4741, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 5.079999999999999e-07, | |
| "loss": 0.4565, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 5.041538461538461e-07, | |
| "loss": 0.4867, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 5.003076923076923e-07, | |
| "loss": 0.4573, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 4.964615384615385e-07, | |
| "loss": 0.4872, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 4.926153846153846e-07, | |
| "loss": 0.4499, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 4.887692307692308e-07, | |
| "loss": 0.4456, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 4.849230769230769e-07, | |
| "loss": 0.437, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 4.810769230769231e-07, | |
| "loss": 0.4494, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 4.772307692307692e-07, | |
| "loss": 0.4497, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 4.733846153846154e-07, | |
| "loss": 0.4457, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 4.695384615384615e-07, | |
| "loss": 0.4571, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 4.6569230769230765e-07, | |
| "loss": 0.4555, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 4.618461538461538e-07, | |
| "loss": 0.4395, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "eval_loss": 0.5783179998397827, | |
| "eval_runtime": 1220.7231, | |
| "eval_samples_per_second": 13.222, | |
| "eval_steps_per_second": 0.414, | |
| "eval_wer": 28.24876127857337, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 4.58e-07, | |
| "loss": 0.483, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.541538461538461e-07, | |
| "loss": 0.5286, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.503076923076923e-07, | |
| "loss": 0.513, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.4646153846153847e-07, | |
| "loss": 0.492, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.426153846153846e-07, | |
| "loss": 0.4849, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.387692307692307e-07, | |
| "loss": 0.4528, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.349230769230769e-07, | |
| "loss": 0.4383, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.3107692307692306e-07, | |
| "loss": 0.4397, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 4.2723076923076923e-07, | |
| "loss": 0.4659, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 4.2338461538461536e-07, | |
| "loss": 0.4536, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 4.1953846153846153e-07, | |
| "loss": 0.4476, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 4.1569230769230765e-07, | |
| "loss": 0.4319, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 4.118461538461538e-07, | |
| "loss": 0.4332, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 4.0799999999999995e-07, | |
| "loss": 0.4134, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 4.041538461538461e-07, | |
| "loss": 0.4494, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 4.003076923076923e-07, | |
| "loss": 0.4804, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 3.9646153846153847e-07, | |
| "loss": 0.4146, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 3.926153846153846e-07, | |
| "loss": 0.4724, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 3.887692307692307e-07, | |
| "loss": 0.4462, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 3.849230769230769e-07, | |
| "loss": 0.4233, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "eval_loss": 0.5746914148330688, | |
| "eval_runtime": 1237.3512, | |
| "eval_samples_per_second": 13.045, | |
| "eval_steps_per_second": 0.408, | |
| "eval_wer": 28.07788276196445, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 3.8107692307692306e-07, | |
| "loss": 0.4978, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 3.7723076923076924e-07, | |
| "loss": 0.4922, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 3.7338461538461536e-07, | |
| "loss": 0.4241, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 3.6953846153846153e-07, | |
| "loss": 0.4194, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 3.656923076923077e-07, | |
| "loss": 0.4515, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 3.618461538461539e-07, | |
| "loss": 0.4537, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 3.5799999999999995e-07, | |
| "loss": 0.4676, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 3.541538461538461e-07, | |
| "loss": 0.5138, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 3.503076923076923e-07, | |
| "loss": 0.4551, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 3.4646153846153847e-07, | |
| "loss": 0.4621, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 3.426153846153846e-07, | |
| "loss": 0.4554, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 3.3876923076923077e-07, | |
| "loss": 0.5044, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 3.3492307692307694e-07, | |
| "loss": 0.4592, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 3.3107692307692306e-07, | |
| "loss": 0.4401, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.272307692307692e-07, | |
| "loss": 0.4405, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.2338461538461536e-07, | |
| "loss": 0.4307, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.1953846153846153e-07, | |
| "loss": 0.3942, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 3.156923076923077e-07, | |
| "loss": 0.4354, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 3.1184615384615383e-07, | |
| "loss": 0.4048, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 3.08e-07, | |
| "loss": 0.4204, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "eval_loss": 0.5712096691131592, | |
| "eval_runtime": 1254.0999, | |
| "eval_samples_per_second": 12.871, | |
| "eval_steps_per_second": 0.403, | |
| "eval_wer": 28.11217672031503, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 3.0415384615384613e-07, | |
| "loss": 0.4269, | |
| "step": 5025 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.0046153846153846e-07, | |
| "loss": 0.4425, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2.966153846153846e-07, | |
| "loss": 0.3973, | |
| "step": 5075 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 2.9276923076923075e-07, | |
| "loss": 0.3994, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 2.8892307692307693e-07, | |
| "loss": 0.4144, | |
| "step": 5125 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 2.8507692307692305e-07, | |
| "loss": 0.4165, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 2.812307692307692e-07, | |
| "loss": 0.4084, | |
| "step": 5175 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 2.773846153846154e-07, | |
| "loss": 0.3953, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 2.7353846153846157e-07, | |
| "loss": 0.3909, | |
| "step": 5225 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 2.6969230769230764e-07, | |
| "loss": 0.3984, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 2.658461538461538e-07, | |
| "loss": 0.3863, | |
| "step": 5275 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 2.62e-07, | |
| "loss": 0.4402, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 2.5815384615384616e-07, | |
| "loss": 0.4355, | |
| "step": 5325 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 2.543076923076923e-07, | |
| "loss": 0.4346, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 2.5046153846153846e-07, | |
| "loss": 0.44, | |
| "step": 5375 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 2.4661538461538463e-07, | |
| "loss": 0.4339, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 2.4276923076923076e-07, | |
| "loss": 0.4363, | |
| "step": 5425 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 2.3892307692307693e-07, | |
| "loss": 0.4694, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 2.3507692307692305e-07, | |
| "loss": 0.4834, | |
| "step": 5475 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 2.3123076923076923e-07, | |
| "loss": 0.4378, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "eval_loss": 0.5695297718048096, | |
| "eval_runtime": 1232.4207, | |
| "eval_samples_per_second": 13.097, | |
| "eval_steps_per_second": 0.41, | |
| "eval_wer": 28.057779407069283, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 2.2738461538461537e-07, | |
| "loss": 0.4635, | |
| "step": 5525 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.2353846153846155e-07, | |
| "loss": 0.4252, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.1969230769230767e-07, | |
| "loss": 0.4115, | |
| "step": 5575 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.1584615384615384e-07, | |
| "loss": 0.4215, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.12e-07, | |
| "loss": 0.4793, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.0815384615384614e-07, | |
| "loss": 0.4638, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 2.043076923076923e-07, | |
| "loss": 0.4225, | |
| "step": 5675 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 2.0046153846153846e-07, | |
| "loss": 0.436, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 1.966153846153846e-07, | |
| "loss": 0.4274, | |
| "step": 5725 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 1.9276923076923076e-07, | |
| "loss": 0.4437, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 1.889230769230769e-07, | |
| "loss": 0.4447, | |
| "step": 5775 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 1.8507692307692308e-07, | |
| "loss": 0.4578, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 1.812307692307692e-07, | |
| "loss": 0.6333, | |
| "step": 5825 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 1.7738461538461538e-07, | |
| "loss": 0.4166, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 1.7353846153846152e-07, | |
| "loss": 0.4178, | |
| "step": 5875 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 1.696923076923077e-07, | |
| "loss": 0.3826, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 1.6584615384615382e-07, | |
| "loss": 0.4209, | |
| "step": 5925 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 1.62e-07, | |
| "loss": 0.4674, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 1.5815384615384614e-07, | |
| "loss": 0.4828, | |
| "step": 5975 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 1.5430769230769232e-07, | |
| "loss": 0.4337, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "eval_loss": 0.5673043727874756, | |
| "eval_runtime": 1224.2663, | |
| "eval_samples_per_second": 13.184, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 27.781653914832727, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.07, | |
| "learning_rate": 1.5046153846153844e-07, | |
| "loss": 0.3946, | |
| "step": 6025 | |
| }, | |
| { | |
| "epoch": 11.07, | |
| "learning_rate": 1.466153846153846e-07, | |
| "loss": 0.4453, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 1.4276923076923076e-07, | |
| "loss": 0.4741, | |
| "step": 6075 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1.389230769230769e-07, | |
| "loss": 0.4201, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1.3507692307692306e-07, | |
| "loss": 0.4456, | |
| "step": 6125 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1.3123076923076923e-07, | |
| "loss": 0.4489, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 1.2738461538461538e-07, | |
| "loss": 0.4399, | |
| "step": 6175 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 1.2353846153846153e-07, | |
| "loss": 0.4538, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 1.196923076923077e-07, | |
| "loss": 0.4297, | |
| "step": 6225 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 1.1584615384615385e-07, | |
| "loss": 0.4568, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 1.12e-07, | |
| "loss": 0.4262, | |
| "step": 6275 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1.0815384615384614e-07, | |
| "loss": 0.4578, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1.043076923076923e-07, | |
| "loss": 0.425, | |
| "step": 6325 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1.0046153846153845e-07, | |
| "loss": 0.4336, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 9.661538461538462e-08, | |
| "loss": 0.3991, | |
| "step": 6375 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 9.276923076923076e-08, | |
| "loss": 0.4253, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 8.892307692307692e-08, | |
| "loss": 0.427, | |
| "step": 6425 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 8.507692307692307e-08, | |
| "loss": 0.428, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 8.123076923076923e-08, | |
| "loss": 0.4389, | |
| "step": 6475 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 7.738461538461538e-08, | |
| "loss": 0.4277, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "eval_loss": 0.565819501876831, | |
| "eval_runtime": 1225.0325, | |
| "eval_samples_per_second": 13.176, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 27.663398886037626, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 7.353846153846153e-08, | |
| "loss": 0.4094, | |
| "step": 6525 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 6.969230769230769e-08, | |
| "loss": 0.4744, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 6.584615384615384e-08, | |
| "loss": 0.4288, | |
| "step": 6575 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 6.2e-08, | |
| "loss": 0.4053, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 5.8153846153846154e-08, | |
| "loss": 0.4049, | |
| "step": 6625 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 5.43076923076923e-08, | |
| "loss": 0.4309, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 5.0461538461538456e-08, | |
| "loss": 0.3928, | |
| "step": 6675 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 4.661538461538461e-08, | |
| "loss": 0.4274, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 4.2769230769230765e-08, | |
| "loss": 0.423, | |
| "step": 6725 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 3.892307692307692e-08, | |
| "loss": 0.4922, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 3.5076923076923074e-08, | |
| "loss": 0.4777, | |
| "step": 6775 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 3.123076923076923e-08, | |
| "loss": 0.4448, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 2.7384615384615383e-08, | |
| "loss": 0.4551, | |
| "step": 6825 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 2.3538461538461538e-08, | |
| "loss": 0.5006, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 1.9692307692307693e-08, | |
| "loss": 0.5345, | |
| "step": 6875 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 1.5846153846153847e-08, | |
| "loss": 0.4159, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 1.2e-08, | |
| "loss": 0.4167, | |
| "step": 6925 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 8.153846153846153e-09, | |
| "loss": 0.4118, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 4.307692307692307e-09, | |
| "loss": 0.3877, | |
| "step": 6975 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 4.615384615384615e-10, | |
| "loss": 0.419, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "eval_loss": 0.5653703808784485, | |
| "eval_runtime": 1226.1688, | |
| "eval_samples_per_second": 13.164, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 27.650982108014144, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "step": 7000, | |
| "total_flos": 2.905221919997952e+19, | |
| "train_loss": 0.5182813051768712, | |
| "train_runtime": 39943.7657, | |
| "train_samples_per_second": 11.216, | |
| "train_steps_per_second": 0.175 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 7000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 2.905221919997952e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |